DESCRIPTION (Taken from application) Obesity is an increasingly prevalent problem in childhood. There is a public health need to disseminate empirically validated treatments for childhood obesity. This proposal is designed to evaluated alternative methods for delivery of a family-based behavioral intervention for an empirically validated treatment of obesity in 8-12 year-old children and their parents. The family-based treatment was designed to maximize efficacy, and includes small groups, didactic teaching, and individual attention that is too labor intensive for widespread clinical and public health use. One of the first steps in development of an intervention for widespread dissemination would be to test a program that can be delivered to families in a more cost effective manner. Previous obesity studies have demonstrated that treatment outcome does not depend upon the intensity of therapist involvement. Research has demonstrated the efficacy and cost- effectiveness of non-didactic, large group interventions for child problems. The present study adapts the large group intervention to obesity treatment. The proposed experiment compares sixty obese children randomized to one of three groups: small group, individualized treatment: large group, non-individualized treatment, and non-treatment control. The primary outcome is the comparisons of changes in percent overweight between the three groups. This study provides the first experimental test of different methods for implementing family-based behavioral interventions for obese children.